Bacterial transport in Gram-negative organisms is initiated by passage of the transported species through a transmembrane beta-barrel in the outer membrane. The transport of iron is particularly important for bacterial growth, and outer membrane iron transporters are therefore major vaccine targets against pathogens such as Neisseria, Haemophilus, and Yersinia. Iron transport across the outer membrane is interesting to study because this family of transporters shows high affinity and specificity for Fe(III)-ligand complexes and because the transport process requires energy derived from the proton motive force across the inner membrane. The required energy is transduced by transient complexation with an integral inner membrane protein, TonB, resulting in a complex that spans both the inner and outer membranes, as well as the periplasmic space. The first crystal structure [1] of an E. coli TonB-dependent receptor, ferric enterobactin receptor (FepA = 80 kDa), revealed that this iron transporter uses a 22-stranded beta-barrel to span the outer membrane, with an unanticipated globular domain folded into the barrel interior. The globular domain appears to function in both ligand binding at the extracellular side of the membrane, and in TonB recognition at the periplasmic side of the membrane. In this 'ground state' structure, the globular domain inside the barrel completely occludes the barrel lumen, and we do not understand how ferric enterobactin (700 Da) is transported. To understand more about iron transport, three projects were initiated in May 2001. (1) Ligand recognition and transport Colicin I receptor (Cir) is a 67 kD Ton-B dependent outer membrane receptor which transports colicins Ia and Ib (~70 kDa) as well as its physiological iron-ligand. Since the structure of FepA does not indicate how a 700 Da molecule should pass through the beta-barrel, we are intrigued by the problem of transporting a protein that is 100 times larger. Colicin I receptor also functions in the uptake of catechol-substituted antibiotics and in the uptake of 2,3-dihydroxybenzoate derivatives. Particularly because colicin I receptor transports catechol-substituted antibiotics, a crystal structure will be useful to the field of antibiotic research. An added benefit from studying this particular receptor is that the structure of colicin Ia has been solved and the domain that should recognise and bind Cir has been identified. We have collected native data for Cir at 2.8A resolution and have introduced 4 additional methionine residues to aid phasing by the multiwavelength anomalous dispersion method. Structure solution is underway. We have also characterized the binding of colicin Ia to Cir by microcalorimetry. The complex has a nanomolar dissociation constant and co-crystallization experiments are in progress. (2) Regulation of iron transport at the E. coli outer membrane Escherichia coli K-12 synthesizes six siderophore-mediated transport systems for the acquisition of Fe3+ (such as colicin Ia and ferric enterobactin receptors, discussed above). Each system encodes a distinct transmembrane receptor for transport across the outer membrane. Transcription of ferric siderophore transport genes can be induced by extracellular ferric citrate. Induction involves FecA, the outer membrane transporter of ferric citrate, FecR, an inner membrane protein that transmits the signal into the cytoplasm, and FecI, a sigma factor that mediates specific binding of the RNA polymerase core enzyme to the promoter region upstream of fecA. We have just solved the structures of FecA alone and in complex with citrate and ferric citrate to resolutions of 2.1A to 3.4A [2]. These three structures together shed new light on apo- and holo-ligand binding. We have deduced the structural mechanism for discrimination between the iron-free and ferric siderophore: the binding of diferric dicitrate, but not iron-free dicitrate alone, causes major conformational rearrangements in the transporter. The structure of FecA bound with iron-free dicitrate represents the first structure of a TonB-dependent transporter bound with an iron-free siderophore. Binding of diferric dicitrate to FecA results in changes in the orientation of the two citrate ions relative to each other and in their interactions with FecA, compared to the binding of iron-free dicitrate. The changes in ligand binding are accompanied by conformational changes in three areas of FecA: two extracellular loops, one plug domain loop and the periplasmic TonB-box motif. The positional and conformational changes in the siderophore and transporter initiate two independent events: ferric citrate transport into the periplasm and transcription induction of the fecABCDE transport genes. From these data we proposed a two-step ligand recognition event: FecA binds iron-free dicitrate in the non-productive state or first step, followed by siderophore displacement to form the transport-competent, diferric dicitrate bound state in the second step. (3) Structure of the Neisseria meningitidis transferrin binding protein complexed to human transferrin Neisseria meningitidis, a Gram-negative bacterium, is the causative agent of bacterial meningitidis. This blood-borne pathogen acquires iron from human transferrin (80 kDa) through an outer membrane transporter complex, transferrin binding proteins A and B (TbpA = 100 kDa; TbpB = 68-85 kDa). TbpA and TbpB form a discrete complex to bind transferrin synergistically, yet each protein is also capable of binding transferrin on its own. Both TbpA and TbpB are current vaccine targets. While TbpB is predicted to be a surface-exposed lipoprotein, TbpA is a member of the TonB-dependent outer membrane receptor/transporter family. We have determined the stoichiometry of the TbpA:TbpB:transferrin complex to be 2:1:1, yielding a total size of 360 kDa. Crystallisation experiments are underway for (a) TbpA alone, (b) TbpA in complex with transferrin, (c) TbpB alone, (d) TbpB in complex with transferrin, and (e) the entire TbpA:TbpB:tranferrin complex. Crystals have been obtained for TbpA complexed with the C-lobe of human transferrin, which diffract to 3.2A resolution. Additional methionine residues have been added to aid phasing. We expect that our crystal structures will aid vaccine development for serotype B meningitis, for which there is currently no vaccine available.